252 PHYSICS 



glass tube over the lower end of which the rubber tube slips, within 

 which is a second glass tube conveying the gas, and which continues 

 down nearly to the merciuy level. The gas enters by this tube, escapes 

 at its lower end, passes up between the two, and goes to the burner 

 through a lateral tube. If the temperature rises too high the mercury 

 column cuts off the supply of gas by the iucreased tension of the vapor ; 

 a by-pass proventiug complete extinction of the flame. The interrupter 

 is attached to a board which slides up and down aloug a graduated 

 scale, by means of which the pressure upon the volatile liquid can be 

 made to balance its tension at any temperature. Using methyl alcohol, 

 for example, its tension at 10° C is SO'""^, at 50° is 382™°^, at 70° is 857"^''i, 

 at 100° is 24115™'". So that with this liquid constant temperatures be. 

 tween these extremes may be had. For other temperatures other liquids 

 more or less volatile than methyl alcohol can be used in its stead. — 

 {J. Phys., viii, 340, October, 1879.) 



Edison made use of his tasimeter for measuring the heat of the solar 

 coroua at the eclipse of July, 1878. In an account subsequently i)ub- 

 lished he gives figures of the instrument and describes the method of 

 using it. A carbon button is inclosed in a heavy metallic semi-conical 

 box. Upon this button rests a metallic i)late holding the lower eud of a 

 strip of vulcanite, the upper end of which is held by a screw which is 

 capable of nice adjustment. When radiant heat is allowed to fall on 

 tlie vulcanite strip, it expands and produces pressure on the carbon, 

 thus reducing its resistance and causing an increase in the strength of 

 any current which may be flowing through it at the time. Ordinarily 

 the tasimeter is made one side of a Wheatstone bridge. To balance it, 

 at Rawlins, a method of "fractional balancing" was devised, which con- 

 sisted of a rheostat formed of an iron wire. No. 24, 4 feet long, wound be- 

 tween two rows of i)ins on a board, used as a shunt around the galva- 

 nometer. When the pins of one row were all connected by a piece of cop- 

 per wire, the resistance of the shunt was so low that only a feeble current 

 passed through the galvanometer, and it was easy to balance it by ad- 

 iusting the tasimeter. The copper wire was then taken off a single pin, 

 increasing the resistance of the shunt by -^ ohm, and the spot of 

 light, deflected nearly ofl" the scale, was again brought to zero. And so 

 on until the whole shunt was removed, and the whole current was used 

 on the galvanometer. Its sensitiveness Avas now adjusted by raisiug the 

 damping magnet until the top of the rod was reached. In this condi- 

 tion it easily indicated the heat of Arcturus focused on the vulcanite by 

 a 4-inch object-glass. But upon the corona the deflection was so violent 

 that the spot of light passed off the scale not to return. — {Am. J. Sci., 

 Ill, xvii, 52, January, 1879.) 



Rid out proposes to show the expansion of glass, as a class experiment, 

 very neatly, by bending a glass tube of small diameter and 18 inches 

 long into the form of a horseshoe, so that the free ends approach each 

 other within a millimeter. Sjjringing these ends apart and introducing 



